Quick question for an opengl guru, I'm in the process of building a render queue and can either reduce the number of shader binding or the number of vertex buffer binding. I just want to know which one is more expensive on newer graphics cards (IE)(opengl >= 3.2)?
How exactly your system performs state changes is not part of the OpenGL specification, so perfomrances may vary betwen diffrent vendors and implementations. The only real way to answer your question is simply by testing it. Write a simple test program (or use the the thing you´re currently working on if possible), which performs alot of buffer bindings to reduce shader bindings and then the opposite. Use a high precision clock to measure the time of both test cases and compare.